COVID-19: The CIDRAP Viewpoint COVID-19: The CIDRAP Viewpoint April 30th, 2020

Part 1: The Future of the COVID-19 : Lessons Learned from Pandemic Kristine A. Moore, MD, MPH Marc Lipsitch, DPhil John M. Barry, MA Michael T. Osterholm, PhD, MPH

Dr. Moore is medical director of the Center for Infectious Research and Policy (CIDRAP). Dr. Lipsitch is the director of the Center for Communicable Disease Dynamics, Department of , Harvard TH Chan School of . John Barry is a professor at Tulane University School of Public Health and Tropical Medicine. Dr. Osterholm is director of CIDRAP, University of Minnesota Regents Professor, and McKnight Presidential Endowed Chair in Public Health.

CIDRAP, founded in 2001, is a global leader in addressing public health preparedness and emerging infectious disease response. Part of the Office of the Vice President for Research (OVPR) at the University of Minnesota, CIDRAP works to prevent illness and death from targeted infectious disease threats through research and the of scientific information into real-world, practical applications, policies, and solutions. For more information, visit: www.cidrap.umn.edu.

COVID-19 Viewpoint reports are made possible with support from the University of Minnesota OVPR and the Bentson Foundation.

COVID-19: The CIDRAP Viewpoint working group: Michael T. Osterholm, PhD, MPH, CIDRAP director Kristine A. Moore, MD, MPH, CIDRAP medical director Julie Ostrowsky, MSc, CIDRAP research associate James Seifert, JD, MS, MPH, CIDRAP program manager Angela Ulrich, PhD, MPH, CIDRAP research associate Alison Kraigsley, PhD, MS, CIDRAP research associate Maya Peters, MPH, CIDRAP program analyst Jim Wappes, CIDRAP editorial director

Editing: Jim Wappes; Report Layout and Design, Cover Design: Hannah Winesburg © 2020 Regents of the University of Minnesota. All rights reserved.

1 Part 1: The Future of the COVID-19 Pandemic: Lessons from Pandemic Influenza Preface Welcome to “COVID-19: The CIDRAP Viewpoint.” We appreciate that other expert groups have produced detailed plans for mitigating SARS-CoV-2 and for reopening the country after stay-at-home orders and other important mitigation steps are eased. Our intent with the Viewpoint is to add key information and address issues that haven’t garnered the attention they deserve and reflect the unique experience and expertise among the CIDRAP team and our expert consultants. We will address timely issues with straight talk and clarity. And the steps we will recommend will be based on our current reality and the best available data. Our goal is to help planners envision some of the situations that might present themselves later this year or next year so that they can take key steps now, while there’s still time. “COVID-19: The CIDRAP Viewpoint” will address such topics as pandemic scenarios going forward, crisis communication, testing, contact tracing, surveillance, supply chains, and epidemiology issues and key areas for research. We will release approximately one to two reports per week. Our hope is that our effort can help you plan more effectively and understand the many aspects of this pandemic more clearly—and for you and your family, friends, and colleagues to be safer. Thank you. – Michael T. Osterholm, PhD, MPH, CIDRAP Director

Introduction When severe acute respiratory syndrome -2 (SARS-CoV-2)—the that causes COVID-19—first emerged in Wuhan, China, in December 2019, even the most experienced international public health experts did not anticipate that it would rapidly spread to create the worst global public health crisis in over 100 years. By January 2020, a few public health officials began sounding the alarm, but it wasn’t until March 11, 2020, that the World Health Organization declared a global pandemic. The virus caught the global community off guard, and its future course is still highly unpredictable; there is no crystal ball to tell us what the future holds and what the “end game” for controlling this pandemic will be. The epidemiology of other serious (SARS-CoV-1, the virus that causes severe acute respiratory syndrome [SARS] and Middle East respiratory syndrome coronavirus [MERS-CoV]) is substantially different from that of SARS-CoV-2; therefore, these do not provide useful models for predicting what to expect with this pandemic. Alternatively, our best comparative model is pandemic influenza. Since the early 1700s, at least eight global influenza have occurred, and four of these occurred since 1900—in 1918-19, 1957, 1968, and 2009- 10. We can potentially learn from past influenza pandemics as we attempt to determine a vision for the future of the COVID-19 pandemic. Identifying key similarities and differences in the epidemiology of COVID-19 and pandemic influenza can help envisioning several possible scenarios for the course of the COVID-19 pandemic. The primary focus of these scenarios is on the temperate Northern Hemisphere, but similar patterns could occur in the Global South, as well. The lack of robust healthcare infrastructure (including a dearth of adequate personal protective equipment) and comorbidities such as other (eg, HIV, TB, ), malnutrition,

2 and chronic in certain areas of the Global South could result in the pandemic being even more severe in those areas, as was noted during the 1918-19 pandemic (Murray 2006).

Epidemiologic Similarities, Pressing Issues Differences Between Covid-19 and 1. Because of a longer , Pandemic Influenza more spread, and a higher Even though coronaviruses are very different from influenza R0, COVID-19 appears to spread more , the COVID-19 pandemic and pandemic influenza easily than flu. share several important similarities. First, SARS-CoV-2 and a 2. A higher R0 means more people will need pandemic influenza virus are novel viral pathogens to which to get infected and become immune before the pandemic can end. the global population has little to no pre-existing , thereby resulting in worldwide susceptibility. Second, SARS- 3. Based on the most recent flu pandemics, CoV-2 and influenza viruses are predominantly spread this outbreak will likely last 18 to 24 months. via the respiratory route by large droplets, but also with a significant component of transmission by smaller . 4. It likely won’t be halted until 60% to 70% of the population is immune. Asymptomatic transmission occurs with both viruses as well, thereby contributing to the spread of each. Finally, both types 5. Depending on control measures and other factors, cases may come in waves of viruses are capable of infecting millions of people and of different heights (with high waves moving rapidly around the globe. signaling major impact) and in different There are also important differences. The first is the intervals. We present 3 possibilities. incubation period; the average incubation period for influenza is 2 days (range, 1 to 4 days); whereas, the average incubation period for COVID-19 is 5 days (range, 2 to 14 days) (Lauer 2020). The longer incubation period for COVID-19 allowed the virus to move silently in different populations before being detected (Kahn 2020, Li 2020). This contributed to an initial environment of complacency before national governments became aware of the severity of the situation. The second important factor is the asymptomatic fraction for the two infections. Although information is still being collected to definitively define the asymptomatic fraction for COVID-19, public health officials have stated that 25% of all cases may be asymptomatic (Rettner 2020) and better serologic studies may revise this percentage upward. A number of studies have explored the asymptomatic fraction for influenza; one review found a pooled mean for the asymptomatic fraction of 16% (range of 4% to 28%) (Leung 2015). Thus, while both viruses can lead to asymptomatic infections, the asymptomatic fraction appears to be somewhat higher for COVID-19 than for influenza. Another consideration is the timeframe of presymptomatic for people who fall ill. One recent study found that the SARS-CoV-2 viral load was highest at the time of symptom onset, suggesting that viral shedding may peak before symptoms occur, leading to substantial presymptomatic transmission (He 2020). A point- study of SARS-CoV-2 in nursing home residents showed that, for 27 residents who were asymptomatic at the time of testing, 24 developed symptoms a median of 4 days later (interquartile range, 3 to 5 days) (Arons 2020), supporting the potential for several days of presymptomatic shedding. For the H1N1 pandemic , one study showed that viral shedding peaks the first 1 to 2 days after symptom

3 onset, suggesting there may be less presymptomatic shedding for pandemic influenza A compared with SARS- CoV-2 (Ip 2016). All of the above factors contribute to viral transmissibility. One way to quantify the transmissibility of a virus is to Recommendations determine the basic reproductive number (R0) for that virus. The R0 is the average number of new infections that 1. States, territories, and tribal health result from a single infected person in a wholly susceptible authorities should plan for the worst-case population (Delamater 2019). The R0 can vary by factors that scenario (Scenario 2), including no availability or . influence the contact rate between people, such as physical distancing strategies and lockdowns aimed at driving the 2. Government agencies and healthcare delivery organizations should develop R0 below 1. An R0 below 1 indicates that that an outbreak is strategies to ensure adequate protection shrinking rather than expanding, since each infected person for healthcare workers when disease is subsequently infecting less than 1 other person. While surges. the R0 is not influenced by herd immunity (which is the 3. Government officials should develop proportion of the population that is immune to a virus), either concrete plans, including triggers for generated by natural or by , immunity reinstituting mitigation measures, for in the population can influence the effective reproductive dealing with disease peaks when they number (RE), which is similar to the R0 but does not depend occur. on having a fully susceptible population (Delamater 2019). 4. Risk communication messaging from Immunity in the population can effectively diminish or end government officials should incorporate an outbreak by driving RE below 1 (Fine 2011). the concept that this pandemic will not be over soon and that people need The R0 for SARS-CoV-2 during the early course of the to be prepared for possible periodic pandemic in China was estimated at 2.0 to 2.5 (Anderson resurgences of disease over the next 2 2020); however, the R0 for SARS-CoV-2 is difficult to years. accurately determine in various geographic regions because of challenges in identifying and testing infected persons, and one study has suggested that the value may be considerably higher (Sanche 2020). Also, for SARS-CoV-2, the R0 is not the same for each person; it can change based on natural variability in viral shedding by infected persons. Even the average value of R0 is not a purely biological quantity—it depends on behavior and contacts. For example, some have speculated that the R0 for SARS-CoV-2 may be higher in areas of denser population or more frequent contacts, such as large cities. In addition, some evidence indicates that some people are “super-spreaders,” as was seen with MERS-CoV and SARS (Frieden 2020, Wong 2015). Some countries appear to have been able to drive their R0 for SARS-CoV-2 below 1 with mitigation measures, although as the mitigation measures are lifted, the R0 in any given area may creep back above 1, leading to disease resurgence over time. The R0 for pandemic influenza has varied by pandemic, but estimates have consistently been around or below 2, suggesting that even past severe viruses have been less transmissible than SARS-CoV-2. A review article published after the 2009-10 pandemic examined a range of studies reporting R0 values for the last four influenza pandemics. While the results varied, the highest median R0 was associated with the 1918 and the 1968 influenza pandemics (both 1.8), followed by the 1957 pandemic (1.65), then the 2009-10 pandemic (1.46); by comparison, seasonal influenza have a median R0 of 1.27 (Biggerstaff 2014).

4 Key Lessons From Past Influenza Pandemics Of eight major pandemics that have occurred since the early 1700s, no clear seasonal pattern emerged for most. Two started in winter in the Northern Hemisphere, three in the spring, one in the summer, and two in the fall (Saunders-Hastings 2016). Seven had an early peak that disappeared over the course of a few months without significant human intervention. Subsequently, each of those seven had a second substantial peak approximately 6 months afterfirst peak. Some pandemics showed smaller waves of cases over the course of 2 years after the initial wave. The only pandemic that followed a more traditional influenza-like seasonal pattern was the 1968 pandemic, which began with a late fall/winter wave in the Northern Hemisphere followed by a second wave the next winter (Viboud 2005). In some areas, particularly in Europe, pandemic-associated mortality was higher the second year. The course of these pandemics was not substantially influenced by a vaccination campaign, except for the 2009- 10 pandemic, during which vaccine initially became available in the about 6 months after the onset of the pandemic, although substantial quantities of vaccine were not available until after the pandemic had peaked in most parts of the country. One report estimated that the vaccination program prevented 700,000 to 1,500,000 clinical cases, 4,000 to 10,000 hospitalizations, and 200 to 500 deaths in the United States (Borse 2013). Following three of the pandemics that have occurred since 1900, the pandemic influenza A gradually became more human-adapted and replaced the predominant seasonally circulating influenza virus to become the main seasonal influenza A strain identified each year. Following the 2009-10 pandemic, the pandemic influenza A H1N1 strain has been co-circulating seasonally along with influenza A H3N2. Key points from observing the epidemiology of past influenza pandemics that may provide insight into the COVID-19 pandemic include the following. First, the length of the pandemic will likely be 18 to 24 months, as herd immunity gradually develops in the human population. This will take time, since limited serosurveillance data available to date suggest that a relatively small fraction of the population has been infected and infection rates likely vary substantially by geographic area. Given the transmissibility of SARS-CoV-2, 60% to 70% of the population may need to be immune to reach a critical threshold of herd immunity to halt the pandemic (Kwok 2020). This may be complicated by the fact that we don’t yet know the duration of immunity to natural SARS-CoV-2 infection (it could be as short as a few months or as long as several years). Based on seasonal coronaviruses, we can anticipate that even if immunity declines after exposure, there may still be some protection against disease severity and reduced contagiousness, but this remains to be assessed for SARS-CoV-2. The course of the pandemic also could be influenced by a vaccine; however, a vaccine will likely not be available until at least sometime in 2021. And we don’t know what kinds of challenges could arise during vaccine development that could delay the timeline. Second, there are several different scenarios for the future of the COVID-19 pandemic, and some of these are consistent with what occurred during past influenza pandemics. These can be summarized as follows and are illustrated in the figure below.

5 ¤ Scenario 1: The first wave of COVID-19 in spring 2020 is followed by a series of repetitive smaller waves that occur through the summer and then consistently over a 1- to 2-year period, gradually diminishing sometime in 2021. The occurrence of these waves may vary geographically and may depend on what mitigation measures are in place and how they are eased. Depending on the height of the wave peaks, this scenario could require periodic reinstitution and subsequent relaxation of mitigation measures over the next 1 to 2 years. ¤ Scenario 2: The first wave of COVID-19 in spring 2020 is followed by a larger wave in the fall or winter of 2020 and one or more smaller subsequent waves in 2021. This pattern will require the reinstitution of mitigation measures in the fall in an attempt to drive down spread of infection and prevent healthcare systems from being overwhelmed. This pattern is similar to what was seen with the 1918-19 pandemic (CDC 2018). During that pandemic, a small wave began in March 1918 and subsided during the summer months. A much larger peak then occurred in the fall of 1918. A third peak occurred during the winter and spring of 1919; that wave subsided in the summer of 1919, signaling the end of the pandemic. The 1957-58 pandemic followed a similar pattern, with a smaller spring wave followed by a much larger fall wave (Saunders-Hastings 2016). Successive smaller waves continued to occur for several years (Miller 2009). The 2009-10 pandemic also followed a pattern of a spring wave followed by a larger fall wave (Saunders-Hastings Figure 1 2016). ¤ Scenario 3: The first wave of COVID-19 in spring 2020 is followed by a “slow burn” of ongoing transmission and case occurrence, but without a clear wave pattern. Again, this pattern may vary somewhat geographically and may be influenced by the degree of mitigation measures in place in various areas. While this third pattern was not seen with past influenza pandemics, it remains a possibility for COVID-19. This third scenario likely would not require the reinstitution of mitigation measures, although cases and deaths will continue to occur. Whichever scenario the pandemic follows (assuming at least some level of ongoing mitigation measures), we must be prepared for at least another 18 to 24 months of significant COVID-19 activity, with hot spots popping up periodically in diverse geographic areas. As the pandemic wanes, it is likely that SARS-CoV-2 will continue to circulate in the human population and will synchronize to a seasonal pattern with diminished severity over time, as with other less pathogenic coronaviruses, such as the betacoronaviruses OC43 and HKU1, (Kissler 2020) and past pandemic influenza viruses have done.

6 References Anderson RM, Heesterbeek H, Klinkenberg D, et al. How will country-based mitigation measures influence the course of the COVID-19 ? Lancet. 2020 Mar 21;395(10228):931-4 [Full text]

Arons, MM, Hatfield, KM, Reddy SC, et al. Presymptomatic SARS-CoV-2 infections and transmission in a skilled nursing facility. N Engl J Med 2020 (published online Apr 24) [Full text]

Biggerstaff M, Cauchemez S, Reed C, et al. Estimates of the reproductive number for seasonal, pandemic, and zoonotic influenza: a systematic review of the literature. BMC Infect Dis 2014 Sep 4;14:480[Full text]

Borse RH, Shrestha SS, Fiore AE, et al. Effects of vaccine program against pandemic influenza A(H1N1) virus, United States, 2009-2010. Emerg Infect Dis 2013 Mar;19(3):439-48 [Full text]

CDC. 1918 Pandemic influenza: three waves. Last updated May 11, 2018; accessed April 27, 2020[Web page]

Delamater PL, Street EJ, Leslie TF, et al. Complexity of the basic reproduction number (R0). Emerg Infect Dis 2019;25(1):1-4 [Full text]

Fine P, Eames K, Heymann DL. “Herd immunity”: a rough guide. Clin Infect Dis 2011;52(7):911-6 [Full text]

Frieden TR, Lee CT. Identifying and interrupting superspreading events—implications for control of severe acute respiratory syndrome coronavirus 2. Emerg Infect Dis 2020 (published online Mar 18) [Full text]

He X, Lau EHY, Wu P, et al. Temporal dynamics in viral shedding and transmissibility of COVID-19. Nat Med 2020 (published online Apr 15) [Full text]

Ip TDKM, Lau LLH, Cowling BJ. The dynamic relationship between clinical symptomatology and viral shedding in naturally acquired seasonal and pandemic influenza virus infections. Clin Infect Dis 2016 Feb 15;62(4):431-7 [Full text]

Kahn R, Peak CM, Fernández-Gracia J, et al. Incubation periods impact the spatial predictability of and outbreaks in Sierra Leone. Proc Natl Acad Sci USA 2020 Mar 3;117(9):5067-73 [Abstract]

Kissler SM, Tedijanto C, Goldstein E, et al. Projecting the transmission dynamics of SARS-CoV-2 through the postpandemic period. Science 2020 (published online Apr 14) [Full text]

Kwok KO, Lai F, Wei WI, et al. Herd immunity—estimating the level required to halt the COVID-19 epidemics in affected countries. J Infect 2020 Mar 21[Full text]

Lauer SA, Grantz KH, Bi Q, et al. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Ann Intern Med 2020 (published online Mar 10) [Full text]

Leung NH, Xu C, Ip DKM, et al. The fraction of influenza virus infections that are asymptomatic: a systematic review and meta-analysis. Epidemiology 2015 Nov;26(6):862-72 [Full text]

Li R, Pei S, Chen B, et al. Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV2) Science 2020 (published online Mar 16) [Full text] 7 Miller MA, Viboud C, Balinska M, et al. The signature features of influenza pandemics — implications for policy. N Engl J Med 2009 Jun 18;360(25):2595-8 [Full text]

Murray CJ, Lopez AD, Chin B, et al. Estimation of potential global pandemic influenza mortality on the basis of vital registry data from the 1918-20 pandemic: a quantitative analysis. Lancet 2006 Dec 23;368(9554):2211-8 [Abstract]

Rettner R. Up to 25% of people with COVID-19 may not show symptoms. Livescience.com [News report]

Sanche S, Lin YT, Xu C, et al. High contagiousness and rapid spread of severe acute respiratory syndrome coronavirus 2. Emerg Infect Dis 2020 Apr 7;26(7) [Full text]

Saunders-Hastings PR, Krewski D. Reviewing the history of pandemic influenza: understanding patterns of emergence and transmission. Pathogens 2016 Dec 6;5(4):66 [Full text]

Viboud C, Grais RF, Lafont BAP, et al. Multinational impact of the 1968 Hong Kong influenza pandemic: evidence for a smoldering pandemic. J Infect Dis 2005 Jul;192(2):233-48 [Full text]

Viboud C, Simonsen L, Fuentes R, et al. Global mortality impact of the 1957–1959 influenza pandemic. J Infect Dis 2016 Mar 1;213(5):738-45 [Full text]

Wong G, Liu W, Liu Y, et al. MERS, SARS, and Ebola: the role of super-spreaders in infectious disease. Cell Microbe 2015;18(4):398-401 [Full text]

8